Method and device for reducing crosstalk interference

Abstract

In a transmission system which makes use of frequency translated signals, said system having a plurality of lines, a method for reducing crosstalk induced on a signal on a first line of said plurality of lines by a signal on a second line of said plurality of lines, comprising estimating a complex coupling factor for the crosstalk, which when multiplied by the signal on the second line estimates the induced crosstalk; and subtracting the estimated induced crosstalk from the signal on the first line. The invention comprises an approximation method for multiplying the complex coupling factor by the signal on the second line that operates on the signal on the second line through pre-rotation, scaling and multiplication by a complex number according to the coupling factor, said complex number being chosen from a bank of predetermined complex numbers so as to obtain the best approximation possible.

What is claimed is:  
     
       1. In a transmission system which makes use of frequency translated signals, particularly a discrete multi tone (DMT) modulated transmission system or an OFDM (Orthogonal Frequency Division Multiplex) transmission system, having a plurality of lines and in which the modulation may be effected using a fast inverse Fourier transform (IFFT), a method for reducing crosstalk interference induced on a signal on a first line of said plurality of lines by a signal on a second line of said plurality of lines, comprising estimating a complex coupling factor for the crosstalk interference, which when multiplied by the signal on the second line estimates the induced crosstalk interference; and subtracting the estimated induced crosstalk interference from the signal on the first line; further comprising multiplying the complex coupling factor by the signal on the second line through an approximation method operating on the signal on the second line including pre-rotation, scaling, and multiplication by a complex number according to the coupling factor, said complex number being chosen from a bank of predetermined complex numbers so as to obtain the best approximation possible. 
     
     
       2. The method as claimed in  claim 1 , comprising providing an amount of complex numbers in the bank in accordance with a maximum acceptable error in the approximation. 
     
     
       3. The method as claimed in  claim 1 , comprising providing the complex numbers in the bank equally spaced. 
     
     
       4. The method as claimed in  claim 1 , comprising providing the complex numbers in the bank unequally spaced. 
     
     
       5. The method as claimed in  claim 1 , comprising providing all the complex numbers in the bank of equal amplitude. 
     
     
       6. The method as claimed in  claim 1 , comprising performing the complex multiplication by the complex number using an iterative vector rotation method. 
     
     
       7. The method as claimed in  claim 6 , comprising performing the complex multiplication by the complex number using the CORDIC vector rotation method. 
     
     
       8. The method as claimed in  claim 1 , comprising performing the pre-rotation by mirroring the signal on the second line in the real and/or the imaginary and/or the 45° axis. 
     
     
       9. The method as claimed in  claim 1 , comprising performing the scaling by multiplying the signal on the second line by a real number. 
     
     
       10. The method as claimed in  claim 9 , wherein the multiplying being preceded by a shift operation. 
     
     
       11. The method as claimed in  claim 9 , comprising choosing the real number so as to obtain the best approximation possible. 
     
     
       12. The method as claimed in  claim 9 , comprising choosing the real number as the amplitude of the coupling factor multiplied by the cosine of the angle mismatch of the chosen complex number as compared with a complex number yielding the exact estimated induced crosstalk interference. 
     
     
       13. The method as claimed in  claim 9 , comprising choosing the real number as the amplitude of the coupling factor. 
     
     
       14. The method as claimed in  claim 9 , comprising choosing the real number from a bank of predetermined real numbers so as to obtain the best approximation possible. 
     
     
       15. The method as claimed in  claim 1 , comprising the further steps of estimating a second different complex coupling factor for a crosstalk interference induced on the signal on the first line by a signal on a third line of said plurality of lines, which when multiplied by the signal on the third line estimates the induced crosstalk interference by the signal on the third line; subtracting the estimated induced crosstalk interference by the signal on the third line from the signal on the first line; and multiplying the second complex coupling factor by the signal on the third line through an approximation method operating on the signal on the third line including pre-rotation, scaling and multiplication by the chosen complex number. 
     
     
       16. The method as claimed in  claim 1 , comprising the further steps of estimating a third different complex coupling factor for a crosstalk interference induced on a signal on a fourth line of said plurality of lines by the signal on the second line, which when multiplied by the signal on the second line estimates the induced crosstalk interference on the signal on the fourth line; subtracting the estimated induced crosstalk interference on the signal on the fourth line from the signal on the fourth line; and multiplying the third complex coupling factor by the signal on the second line through an approximation method operating on the signal on the second line including pre-rotation, scaling and multiplication by the chosen complex number. 
     
     
       17. In a transmission system which makes use of frequency translated signals, particularly a discrete multi tone (DMT) modulated transmission system or an OEDM (Orthogonal Frequency Division Multiplex) transmission system, in which the modulation may be effected using a fast inverse Fourier transform (IFFT), a method for reducing crosstalk interference induced on a signal S N  on a first line N by signals D 1 , . . . , D N−1 , each on a respective line  1 , . . . , N−1, comprising the steps of: 
       (i) associating a coupling factor α′ 1     —     N , . . . , α′ N−1     —     N  with the respective line  1 , . . . , N−1, said coupling factor α′ j     —     N , 1≦j≦N−1, being a complex number and when multiplied by the signal D j , 1≦j≦N−1, on its associated line j, 1≦j≦N−1, estimating the crosstalk interference I j     —     N  induced on the signal S N  on the first line N by the signal D j  on its associated line j;  
       (ii) reducing the crosstalk interference I N  on the signal S N  on the first line N by subtracting an estimated crosstalk interference I′ N  from the signal S N  on the first line N, said estimated crosstalk interference I′ N  being computed from said coupling factors α′ 1     —     N , . . . , α′ N−1     —     N , and said signals D 1 , . . . , D N−1  on the respective line  1 , . . . , N−1;  
       (iii) pre-rotating and scaling each of the signals D 1 , . . . , D N−1  on the respective line  1 , . . . , N−1 according to the coupling factor α′ 1     —     N , . . . , α′ N−1     —     N  associated with the respective line;  
       (iv) summing all of the pre-rotated and scaled signals D* 1 , . . . , D* N−1 , obtained in (iii); and  
       (v) multiplying the sum ΣD* obtained in (iv) by a single complex number β N , which product is used as the estimated crosstalk interference I′ N  in step (ii).  
     
     
       18. The method as claimed in  claim 17 , comprising performing the pre-rotating of each of the signals D 1 , . . . , D N−1  on the respective line  1 , . . . , N−1 according to the coupling factor α′ 1     —     N , . . . , α′ N−1     —     N  associated with the respective line by mirroring each of the signals D 1 , . . . , D N−1  in the real and/or the imaginary and/or the 45° axis. 
     
     
       19. The method as claimed in  claim 17 , comprising performing the multiplying of the sum ΣD* by the single complex number β N  by using an iterative vector rotation method, particularly the CORDIC vector rotation method. 
     
     
       20. The method as claimed in  claim 17 , comprising performing the scaling of each of the signals D 1 , . . . , D N−1  on the respective line  1 , . . . , N−1 according to the coupling factor α′ 1     —     N , . . . , α′ N−1     —     N  associated with the respective line by multiplying each of the signals D 1 , . . . , D N−1  by respective real numbers chosen from a bank of predetermined real numbers so as to obtain the best approximation possible. 
     
     
       21. In a transmission system which makes use of frequency translated signals, particularly a discrete multi tone (DMT) modulated transmission system or an OFDM (Orthogonal Frequency Division Multiplex) transmission system, in which the modulation may be effected using a fast inverse Fourier transform (IFFT), a method for reducing crosstalk interferences induced by a signal D N  on a first line N on signals S 1 , . . . , S N−1  on a respective line  1 , . . . , N−1, comprising the steps of: 
       (i) associating coupling factors α′ N     —     1 , . . . , α′ N     —     N−1 , . . . , α′ N     —     N−1  with the first line N, each of said coupling factors being a complex number and when multiplied by the signal D N  on the first line N estimating the respective crosstalk interference I N     —     1 , . . . , I N     —     N−1  induced on each of the signals S 1 , . . . , S N−1  on the respective line  1 , . . . , N−1 by the signal D N  on the first line N;  
       (ii) reducing the respective crosstalk interference I N     —     1 , . . . , I N     —     N−1  on each of the signals S 1 , . . . , S N−1  on the respective line  1 , . . . , N−1 by subtracting a respective estimated crosstalk interference I′ N     —     1 , . . . , I′ N     —     N−1  from each of the signals S 1 , . . . , S N−1  on the respective line  1 , . . . , N−1, said respective estimated crosstalk interference I′ N     —     1 , . . . , I′ N     —     N−1  being computed from the respective coupling factor α′ N     —     1 , . . . , α′ N     —     N−1 , and the signal D N  on the first line N;  
       (iii) multiplying the signal D N  on the first line N by a single complex number β N  yielding a product D** N ;  
       (iv) replicating the product D** N  for obtaining N−1 equal products D** N     —     1 , . . . , D** N     —     N−1 ;  
       (v) pre-rotating and scaling the respective product D** N     —     1 , . . . , D** N     —     N−1  according to the respective coupling factor α′ N     —     1 , . . . , α′ N     —     N−1 ; and  
       (vi) using the respective pre-rotated and scaled products obtained in (v) as the respective estimated crosstalk interference I′ N     —     1 , . . . , I′ N     —     N−1  in step (ii).  
     
     
       22. The method as claimed in  claim 21 , comprising performing the pre-rotating of the respective product D** N     —     1 , . . . , D** N     —     N−1  according to the respective coupling factor α′ N     —     1 , . . . , α′ N     —     N−1  by mirroring the respective product D** N     —     1 , . . . , D** N     —     N−1  in the real and/or the imaginary and/or the 45° axis. 
     
     
       23. The method as claimed in  claim 21 , comprising performing the multiplying of the signal D N  on the first line N by the single complex number β N  by using an iterative vector rotation method, particularly the CORDIC vector rotation method. 
     
     
       24. The method as claimed in  claim 21 , comprising performing the scaling of the respective product D** N     —     1 , . . . , D** N     —     N−1  according to the respective coupling factor α′ N     —     1 , . . . , α′ N     —     N−1 , and by multiplying the respective product D** N     —     1 , . . . , D** N     —     N−1  by a respective real number chosen from a bank of predetermined real numbers so as to obtain the best approximation possible. 
     
     
       25. In a transmission system which makes use of frequency translated signals, particularly a discrete multi tone (DMT) modulated transmission system or an OFDM (Orthogonal Frequency Division Multiplex) transmission system, in which the modulation may be effected using a fast inverse Fourier transform (IFFT), a device for reducing crosstalk interference provided with means for performing the method as claimed in  claim 1 . 
     
     
       26. In a transmission system which makes use of frequency translated signals, particularly a discrete multi tone (DMT) modulated transmission system or an OFDM (Orthogonal Frequency Division Multiplex) transmission system, in which the modulation may be effected using a fast inverse Fourier transform (IFFT), a transceiver comprising the device as claimed in  claim 25 . 
     
     
       27. A transmission system which makes use of frequency translated signals, particularly a discrete multi tone (DMT) modulated transmission system or an OFDM (Orthogonal Frequency Division Multiplex) transmission system, in which the modulation may be effected using a fast inverse Fourier transform (IFFT), comprising the device as claimed in  claim 25 . 
     
     
       28. A transmission system which makes use of frequency translated signals, particularly a discrete multi tone (DMT) modulated transmission system or an OFDM (Orthogonal Frequency Division Multiplex) transmission system, in which the modulation may be effected using a fast inverse Fourier transform (IFFT), comprising the the transceiver as claimed in  claim 26 .